This study aims to analyze the changes in urban spatial perception across different time periods and geographical spaces, as well as to explore the visual elements affecting these perceptual changes. By examining street view images of Shanghai in 2013 and 2019, the research investigates the distribution of urban spatial perception, the geographical distribution of perceptual enhancement, and the correlation analysis among perceptions.
Key findings include:
Urban spatial perception improvement is mainly concentrated in the peripheral areas of the city, such as Baoshan District and Pudong District, with the most significant enhancement in the areas near the Huangpu River in Pudong District. At the same time, there are instances of perceptual decline in the city center areas, like Putuo District, Changning District, and Xuhui District. Through the statistical and visualization analysis of perceptual differences, it is found that the overall distribution of urban perceptual difference scores follows a normal distribution. Although there are areas with perceptual score decline in the city, the proportion of perceptual enhancement is higher, indicating an overall trend of increasing perception in urban development. From the inner city to the outer city, there is a consistency in perceptual enhancement across urban spaces. This reflects the city's construction strategy during these six years, with more focus on the central areas between the inner and outer cities. Through correlation analysis, it is found that the differences in trees, roads, and plants have a positive correlation with perceptual differences, while the differences in buildings, sky, sidewalks, walls, and fences have a negative correlation. In urban planning and construction, it is essential to fully consider the balance and coordination of various visual elements to improve the perception level of urban residents. This research offers valuable insights into the changes in urban spatial perception for urban planners and builders, helping them better understand the perceptual needs of city residents and optimize urban construction strategies.
The research is carried out under the SHAPC-lab and the lab director & the corresponding author:Jie He academic social networks page:
http://faculty.hitsz.edu.cn/hejie
For more information related to the research, please follow the laboratory's WeChat Official Account:
- MIT PlacePulse 2.0
- Jupyter notebook
- ArcGIS software
- Pillow
- Pandas
- Numpy
- Matplotlib
- sklearn
- training model and drawing figure source code
processing_code
- Street View image perception scoring
streetview_data
- Urban perception and street view collection point data
gis_data